Fig 1.
Fibronectin induces faster migration speed in 2D and 3D substrates.
(A) Effects of different 2D substrates on migration speed (24h) of Linv/HE-cad (Cal27) or Hinv/LE-cad (SCC25) OSCC cell lines (n = 3); (B-E) Cell migration trajectory of Linv/HE-cad (B-C) or Hinv/LE-cad (D-E) cells plated on laminin (B and D) or fibronectin (C and E); (F) Effects of different 3D substrates on migration speed of Linv/HE-cad and Hinv/LE-cad cell lines (n = 3). Results are expressed as mean ± SEM. (*) p<0.05 according to One-way analysis of variance (ANOVA) followed by Tukey’s post-test.
Fig 2.
RhoGTPase activation varies according to extracellular matrix composition and tumor differentiation levels.
FRET analysis and pull down assay for Rac1 (A) and RhoA (B) of Linv/HE-cad (Cal27) or Hinv/LE-cad (SCC25) OSCC plated in laminin (2μg/ml) or fibronectin (2μg/ml). Raichu-Rac1-V12 and Raichu-RhoA-Q63L represents the constitutively activated isoform. Results are expressed as mean ± SD. (*) p<0.05, n = 4.
Fig 3.
Decreased cell-cell and increased cell-ECM adhesion proteins characterize invasive OSCC.
Representative western blotting images of cell-cell (E-cadherin, N-cadherin), cell-ECM (paxillin, vinculin and FAK) and integrins (α4, α5, αv, β1 and β3) from Linv/HE-cad (Cal27) or Hinv/LE-cad (SCC25) OSCC total cell lysates. Densitometry values for each protein were normalized to the loading control (β-Tubulin).
Fig 4.
Fibronectin induces smaller adhesion on low E-cadherin expression OSCC cell line.
Linv/HE-cad (A) or Hinv/LE-cad (B) invasive OSCC were plated on laminin or fibronectin, fixed and stained for E-cadherin and actin, paxillin, vinculin and FAK. White arrows indicate the signal of E-cadherin between cells. Scale bar = 20μm. Data regarding adhesion properties (C) were obtained using Total Internal Reflectance Fluorescent microscopy analysis of Hinv/LE-cad OSCC cells expressing paxillin-GFP and plated on laminin (light gray) or fibronectin (dark gray). The data shows the assembly speed (μm/sec), adhesion area (μm2), total adhesion area (as % of total protrusion area) and adhesion length (μm). Results are expressed as mean ± SEM. (*) p = 0.05; (**) p < 0.01, according to Student T—test.
Fig 5.
Human oral squamous cell carcinoma biopsies show differential distribution of adhesion proteins between center of the tumor cells and tumor-adjacent epithelia.
Regions of biopsies corresponding to the epithelia adjacent to the tumor (A) and from the center of the tumor (B) were submitted to immunostaining for E-cadherin, paxillin, vinculin or FAK (green) and actin staining (magenta). Inserts demonstrated in actin staining, were digitally magnified (5x) to show intracellular localization. Representative images from different patients (n = 10), scale bar = 50μm or 20μm.
Fig 6.
Effects of the differential composition of extracellular matrix on cell adhesion and signaling of Oral Squamous Cell Carcinoma.
OSCC with high E-cadherin levels (blue cells) shows collective and single cell migration in the presence of laminin and collective non-directional migration in fibronectin. This switch correlated to an increase in Rac1 and a decrease on RhoA activation and modulation of the vinculin levels in adhesion, induced by the fibronectin-enriched environment. For OSCCs with low E-cadherin levels (orange cells), fibronectin induced smaller adhesions and increased Rac1 signaling, which correspond to a fast single cell migration phenotype. This model proposes that the ECM composition can trigger the tumor invasive behavior according to differentiation levels of OSCC cells.